A primer of ecological statistics /

Includes bibliographical indexes.

Contents
PART I
Fundamentals of
Probability and
Statistical Thinking
CHAPTER 1
An Introduction to
Probability 3
What Is Probability? 4
Measuring Probability 4
The Probability of a Single Event:
Prey Capture by Carnivorous Plants 4
Estimating Probabilities by Sampling 7
Problems in the Definition of Probability 9
The Mathematics of Probability 11
Defining the Sample Space 11
Complex and Shared Events: Combining
Simple Probabilities 13
Probability Calculations:Milkweeds and
Caterpillars 15
Complex and Shared Events: Rules for
Combining Sets 18
Conditional Probabilities 21
Bayes’ Theorem 22
Summary 24
CHAPTER 2
Random Variables and
Probability Distributions 25
Discrete Random Variables 26
Bernoulli Random Variables 26
An Example of a Bernoulli Trial 27
Many Bernoulli Trials = A Binomial Random
Variable 28
The Binomial Distribution 31
Poisson Random Variables 34
An Example of a Poisson Random Variable:
Distribution of a Rare Plant 36
The Expected Value of a Discrete Random
Variable 39
The Variance of a Discrete Random Variable 39
Continuous Random Variables 41
Uniform Random Variables 42
The Expected Value of a Continuous Random
Variable 45
Normal Random Variables 46
Useful Properties of the Normal
Distribution 48
Other Continuous Random Variables 50
The Central Limit Theorem 53
Summary 54
CHAPTER 3
Summary Statistics: Measures
of Location and Spread 57
Measures of Location 58
The Arithmetic Mean 58
Other Means 60
Other Measures of Location: The Median and
the Mode 64
When to Use Each Measure of Location 65
Measures of Spread 66
The Variance and the Standard Deviation 66
The Standard Error of the Mean 67
Skewness, Kurtosis, and Central Moments 69
Quantiles 71
Using Measures of Spread 72
Some Philosophical Issues Surrounding
Summary Statistics 73
Confidence Intervals 74
Generalized Confidence Intervals 76
Summary 78
CHAPTER 4
Framing and Testing
Hypotheses 79
Scientific Methods 80
Deduction and Induction 81
Modern-Day Induction: Bayesian Inference 84
The Hypothetico-Deductive Method 87
Testing Statistical Hypotheses 90
Statistical Hypotheses versus Scientific
Hypotheses 90
Statistical Significance and P-Values 91
Errors in Hypothesis Testing 100
Parameter Estimation and Prediction 104
Summary 105
CHAPTER 5
Three Frameworks for
Statistical Analysis 107
Sample Problem 107
Monte Carlo Analysis 109
Step 1: Specifying the Test Statistic 111
Step 2: Creating the Null Distribution 111
Step 3: Deciding on a One- or Two-Tailed
Test 112
Step 4: Calculating the Tail Probability 114
Assumptions of the Monte Carlo Method 115
Advantages and Disadvantages of the Monte
Carlo Method 115
Parametric Analysis 117
Step 1: Specifying the Test Statistic 117
Step 2: Specifying the Null Distribution 119
Step 3: Calculating the Tail Probability 119
Assumptions of the Parametric Method 120
Advantages and Disadvantages of the
Parametric Method 121
Non-Parametric Analysis: A Special Case of
Monte Carlo Analysis 121
Bayesian Analysis 122
Step 1: Specifying the Hypothesis 122
Step 2: Specifying Parameters as Random
Variables 125
Step 3: Specifying the Prior Probability
Distribution 125
Step 4: Calculating the Likelihood 129
Step 5: Calculating the Posterior Probability
Distribution 129
Step 6: Interpreting the Results 130
Assumptions of Bayesian Analysis 132
Advantages and Disadvantages of Bayesian
Analysis 133
Summary 133
PART II
Designing Experiments
CHAPTER 6
Designing Successful
Field Studies 137
What Is the Point of the Study? 137
Are There Spatial or Temporal Differences in
Variable Y? 137
What Is the Effect of Factor X on
Variable Y? 138
Are the Measurements of Variable Y Consistent
with the Predictions of Hypothesis H? 138
Using the Measurements of Variable Y,
What Is the Best Estimate of Parameter θ
in Model Z? 139
Manipulative Experiments 139
Natural Experiments 141
Snapshot versus Trajectory Experiments 143
The Problem of Temporal Dependence 144
Press versus Pulse Experiments 146
Replication 148
How Much Replication? 148
How Many Total Replicates Are Affordable? 149
The Rule of 10 150
Large-Scale Studies and Environmental
Impacts 150
Ensuring Independence 151
Avoiding Confounding Factors 153
Replication and Randomization 154
Designing Effective Field Experiments and
Sampling Studies 158
Are the Plots or Enclosures Large Enough to
Ensure Realistic Results? 158
What Is the Grain and Extent of the Study? 158
Does the Range of Treatments or Census
Categories Bracket or Span the Range of
Possible Environmental Conditions? 159
Have Appropriate Controls Been Established
to Ensure that Results Reflect Variation Only
in the Factor of Interest? 160
Have All Replicates Been Manipulated in the
Same Way Except for the Intended
Treatment Application? 160
Have Appropriate Covariates Been Measured
in Each Replicate? 161
Summary 161
CHAPTER 7
A Bestiary of Experimental
and Sampling Designs 163
Categorical versus Continuous Variables 164
Dependent and Independent Variables 165
Four Classes of Experimental Design 165
Regression Designs 166
ANOVA Designs 171
Alternatives to ANOVA: Experimental
Regression 197
Tabular Designs 200
Alternatives to Tabular Designs: Proportional
Designs 203
Summary 204
CHAPTER 8
Managing and Curating
Data 207
The First Step:Managing Raw Data 208
Spreadsheets 208
Metadata 209
The Second Step: Storing and Curating the
Data 210
Storage: Temporary and Archival 210
Curating the Data 211
The Third Step: Checking the Data 212
The Importance of Outliers 212
Errors 214
Missing Data 215
Detecting Outliers and Errors 215
Creating an Audit Trail 223
The Final Step: Transforming the Data 223
Data Transformations as a Cognitive Tool 224
Data Transformations because the Statistics
Demand It 229
Reporting Results: Transformed or Not? 233
The Audit Trail Redux 233
Summary: The Data Management Flow
Chart 235
CHAPTER 9
Regression 239
Defining the Straight Line and Its Two
Parameters 239
Fitting Data to a Linear Model 241
Variances and Covariances 244
Least-Squares Parameter Estimates 246
Variance Components and the Coefficient of
Determination 248
Hypothesis Tests with Regression 250
The Anatomy of an ANOVA Table 251
Other Tests and Confidence Intervals 253
Assumptions of Regression 257
Diagnostic Tests For Regression 259
Plotting Residuals 259
Other Diagnostic Plots 262
The Influence Function 262
Monte Carlo and Bayesian Analyses 264
Linear Regression Using Monte Carlo
Methods 264
Linear Regression Using Bayesian Methods 266
Other Kinds of Regression Analyses 268
Robust Regression 268
Quantile Regression 271
Logistic Regression 273
Non-Linear Regression 275
Multiple Regression 275
Path Analysis 279
Model Selection Criteria 282
Model Selection Methods for Multiple
Regression 283
Model Selection Methods in Path Analysis 284
Bayesian Model Selection 285
Summary 287
CHAPTER 10
The Analysis of Variance 289
Symbols and Labels in ANOVA 290
ANOVA and Partitioning of the Sum of
Squares 290
The Assumptions of ANOVA 295
Hypothesis Tests with ANOVA 296
Constructing F-Ratios 298
A Bestiary of ANOVA Tables 300
Randomized Block 300
Nested ANOVA 302
Two-Way ANOVA 304
ANOVA for Three-Way and n-Way Designs 308
Split-Plot ANOVA 308
Repeated Measures ANOVA 309
ANCOVA 314
Random versus Fixed Factors in ANOVA 317
Partitioning the Variance in ANOVA 322
After ANOVA: Plotting and Understanding
Interaction Terms 325
Plotting Results from One-Way ANOVAs 325
Plotting Results from Two-Way ANOVAs 327
Understanding the Interaction Term 331
Plotting Results from ANCOVAs 333
Comparing Means 335
A Posteriori Comparisons 337
A Priori Contrasts 339
Bonferroni Corrections and the Problem of
Multiple Tests 345
Summary 348
CHAPTER 11
The Analysis of Categorical
Data 349
Two-Way Contingency Tables 350
Organizing the Data 350
Are the Variables Independent? 352
Testing the Hypothesis: Pearson’s Chi-square
Test 354
An Alternative to Pearson’s Chi-Square:
The G-Test 358
The Chi-square Test and the G-Test for R × C
Tables 359
Which Test To Choose? 363
Multi-Way Contingency Tables 364
Organizing the Data 364
On to Multi-Way Tables! 368
Bayesian Approaches to Contingency Tables 375
Tests for Goodness-of-Fit 376
Goodness-of-Fit Tests for Discrete
Distributions 376
Testing Goodness-of-Fit for Continuous
Distributions: The Kolmogorov-Smirnov
Test 380
Summary 382
CHAPTER 12
The Analysis of Multivariate
Data 383
Approaching Multivariate Data 383
The Need for Matrix Algebra 384
Comparing Multivariate Means 387
Comparing Multivariate Means of Two
Samples: Hotelling’s T2 Test 387
Comparing Multivariate Means of More Than
Two Samples: A Simple MANOVA 390
The Multivariate Normal Distribution 394
Testing for Multivariate Normality 396
Measurements of Multivariate Distance 398
Measuring Distances between Two
Individuals 398
Measuring Distances between Two Groups 402
Other Measurements of Distance 402
Ordination 406
Principal Component Analysis 406
Factor Analysis 415
Principal Coordinates Analysis 418
Correspondence Analysis 421
Non-Metric Multidimensional Scaling 425
Advantages and Disadvantages of Ordination
427
Classification 429
Cluster Analysis 429
Choosing a Clustering Method 430
Discriminant Analysis 433
Advantages and Disadvantages of
Classification 437
Multivariate Multiple Regression 438
Redundancy Analysis 438
Summary 444
CHAPTER 13
The Measurement of Biodiversity
449
Estimating Species Richness 450
Standardizing Diversity Comparisons through
Random Subsampling
Rarefaction Curves: Interpolating Species
Richness 455
The Expectation of the Individual-Based Rarefaction
Curve 459
Sample-Based Rarefaction Curves:Massachusetts
Ants 461
Species Richness versus Species Density 465
The Statistical Comparison of Rarefaction
Curves 466
Assumptions of Rarefaction 467
Asymptotic Estimators: Extrapolating
Species Richness 470
Rarefaction Curves Redux: Extrapolation and
Interpolation 476
Estimating Species Diversity and Evenness
476
Hill Numbers 479
Software for Estimation of Species Diversity
481
Summary 482
CHAPTER 14
Detecting Populations
and Estimating their Size 483
Occupancy 485
The Basic Model: One Species, One Season,
Two Samples at a Range of Sites 487
Occupancy of More than One Species 493
A Hierarchical Model for Parameter Estimation
and Modeling 495
Occupancy Models for Open Populations 501
Dynamic Occupancy of the Adelgid in Massachusetts
505
Estimating Population Size 506
Mark-Recapture: The Basic Model 507
Mark-Recapture Models for Open Populations
516
Occupancy Modeling and Mark-Recapture:
Yet More Models 518
Sampling for Occupancy and Abundance
519
Software for Estimating Occupancy and
Abundance 521
Summary 522
APPENDIX
Matrix Algebra for Ecologists 523
Glossary 535
Literature Cited 565
Index 583